The research proposed in this grant is expected to contribute to understanding mechanisms of differentiation of eye tissues, with particular emphasis on cornea and lens. We describe a gene program for epithelial- mesenchymal transformation (EMT) in normal development that we believe is turned on abnormally in certain optic pathologies. This gene program may be turned on in part for motility (as in primary mesenchyme) or in whole to create true fibroblasts (as in secondary mesenchyme). The outline of the work is as follows. I.Epithelial-mesenchymal transformations contributing to normal eye development. The cephalic neural crest arises by epithelial-mesenchymal transformation from the cranial neural folds. We will use beta- galactosidase expressing, replication incompetent retroviruses to label cells to find out if one group of crest cells (more like secondary mesenchyme) unique to the mouse or chicken head gives rise to corneal fibroblasts and another (more like trunk crest) to corneal nerves and conjuntival melanocytes. Also, we will learn the mesenchyme of origin of the corneal endothelium. Using immunohistochemistry and in situ hybridization, we will look for expression of genes for molecules that might cause EMT: CAMs (cell adhesion molecules), SAMs (substrate adhesion molecules), protooncogenes. II.Tissue phenotype transitions by corneal fibroblasts and epithelia. We expect to be able to transform differentiated corneal fibroblasts to epithelioid cells by transfection with cDNAs for E-cadherin, alpha6beta1 integrin, and syndecan, and to transform differentiated corneal epithelial and endothelial cells to fibroblasts by transfection with c-src and c-mos DNAs, and antibodies to cadherins and syndecan or their antisense DNA. We will examine the transformed cells to assay whether parts or all of the tissue phenotype is turned on or off by a given gene. III.Epithelial-mesenchymal transformation from lens and cornea in collagen gels. We will examine by the retrovirus-label method the developmental potentials of the mesenchymal cells that form from lens and corneal epithelia in gels. Since they look like secondary mesenchyme, we expect they will form cartilage. CAMs, SAMs, and protooncogenes will be examined in the lens epithelium before and after it is induced to give rise to mesenchyme in collagen gels. We will use subtractive hybridization to search for the mesenchymal master gene(s) that is turned on during collagen-induced transformation of lens to fibroblasts and compare its sequence(s) with those of CAMs, SAMs, protooncogenes, Homeobox genes, growth factors, and other candidates to be master genes. IV. Epithelial-mesenchymal transformation in eye pathologies. We describe relevant CAMs, protooncogenes, and master genes in preinvasive and invasive conjunctival tumors, and metaplasias of retinal pigmented and nonpigmented epithelia. We investigate the interesting possibility that the gene program for metastasis is an abnormally reactivated embryonic program for epithelial-mesenchymal transformation.